Tower Semiconductor Ltd. (TSEM) Q3 2014 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by. Welcome to the TowerJazz Third Quarter 2014 Results Conference Call. [Operator Instructions] As a reminder, this conference is being recorded, November 13, 2014. Joining us today are Mr. Russell Ellwanger, TowerJazz's CEO; and Mr. Oren Shirazi, CFO. I would now like to turn the conference over to Ms. Limor Silberberg. Ms. Silberberg, please go ahead.

Thank you, and welcome to TowerJazz financial results conference call for the third quarter of 2014. Before we begin, I would like to remind you that some statements made during this call may be forward-looking and are subjected to uncertainties and risk factors that could cause actual results to be different from those currently expected. These uncertainties and risk factors are fully disclosed in our Form 20-F, F-4, F-3 and 6-K filed with the Securities and Exchange Commission as well as filed with the Israeli Securities Authorities. They are also available on our website. TowerJazz assumes no obligation to update any such forward-looking statements. Now I would like to turn the call over to our CEO, Mr. Russell Ellwanger. Russell, please go ahead.

Thank you, Limor. Welcome to all of you to our third quarter 2014 results conference call. Q3 was a very strong and exciting quarter, with 70% year-over-year growth. I'll first comment on non-Panasonic core, or what we define as organic growth. As a continuation to the core organic-related growth we reported in the second quarter, our top 10 customers by revenue presented a growth of 33% versus the revenue levels for the same customers 1 year earlier, and 34% year-to-date. Also quite significant is the fact that all business units are represented within this top 10 customer list. We have mentioned several times that the number of new masks entering our factories is a very strong indicator of growth from 1 to 3 years out. Q3 new masks entering our factories was 6,100, 11% higher than Q2 '14, which quarter itself is a record quarter in this metric, and Q3 was 30% higher than Q3 of 2013. Also, year-to-date, masks entering our factories are 32% higher than the same period in 2013. Q3 and year-to-date design wins were also at record highs for the company. Growth numbers and future indicators like this can only be achieved by serving customers who trust us with a major portion of their business, and by having these customers being within strong industry growth segments. We expect these growth numbers and indicators to continue. As stated, overall Q3 corporate revenue grew 70% year-over-year, specific to TowerJazz Panasonic semiconductor company or TPSCo, during the third quarter, we won 5 new projects. These projects are projected to bring greater than $40 million of new revenue in 2016, and greater than $60 million in 2017. This is on top of the Q2 activities we previously reported. If we combine Q2 and Q3 activities, the projection of about -- is about $100 million for 2016 and about $140 million for 2017 of third-party foundry revenues be manufactured in the TPSCo factories. Panasonic remains a stable and supportive partner in the growth of the TPSCo venture, with multiple new designs taping out. We look forward to continue the excellent collaboration already established. The open and trusting relationship has yielded a venture that has produced no negative surprises since its April 1 inception. Now leaving TPSCo-specific performance I'll move to our overall business unit activities, each one seeing strong positive momentum, as we move through 2014. The CMOS image sensor business unit continues to be strong in core TowerJazz activities. And with the addition of TPSCo, we have been continuing to generate new businesses within several areas that we previously did not serve. In the high-end photography space, we see an interesting trend in the market. The standard digital-still camera and video camera markets are shrinking quickly, simply due to the availability of equivalent quality of still or video photography in high-end smartphones, while, as well a new market of mirror-less professional cameras is emerging. The availability of high-end large sensors of 1 inch and 4/3 inch with very fast output allows true DSLR quality, with interchangeable lenses on Blauw [ph] lightweight small-size cameras. In addition, the fast-growing market of 4k super high-definition LED screens is fueling this market. The new mirror-less cameras not only provide DSLR quality still images, but also allow for high-end video and high-definition or super high-definition quality. Due to this trend and the acknowledged leading position of Panasonic in this market, we expect continued strong growth in our TPSCo high-end sensor market. The 3D gesture control market is also a driver for our products. It is fast growing and becoming a standard, not only in gaming consoles, such as the Microsoft Kinect, but also in mobile PCs and tablets. There have been multiple recent releases demonstrating astonishing applications enabled by PC 3D camera capabilities. Our technology and leading position in this market allows us to grow substantially, and we will see continuous high growth, specifically, in this market, serving market leaders as our main customers. We expect to say more after the January CES conference in Las Vegas. Our TOPS business unit is one in which we transfer customer proprietary process flows to our factories, driven predominantly by integrated device makers, wishing to either close their existing factories that maybe have non-optimal utilizations or who are adapting a satellite model, where their growth is moved into foundries, rather than themselves investing in additional fab infrastructures. Within the past quarter, we had 40 new customer products entering into our factories from this TOPS business unit, and as well experienced our first production ramp at TPSCo in a record time of 7 months. At TPSCo, we have active engagements with 4 customers within this TOPS business model. In our mixed signal CMOS business unit, we continue to serve customers with strong technology platforms, enriched with additional devices and customized process variance to enable our customers to penetrate new markets. We're seeing increased demand for low power and low-leakage devices to serve applications, such as sensors for Internet of Things. The Internet of Things sensor market is increasing in significance and requires an expansion of the current performance envelope, including ultra-low leakage devices and IT blocks. We are addressing these needs and released a dedicated set of IP blocks. In addition, we received an international certification for new enhanced security protocol necessary for advanced products for one of our key customers. Within mixed signal CMOS, at TPSCo, we've released 65-nanometer and 45-nanometer foundry-type process design kits to selected customers that have started designing products for digital, analog and RF CMOS applications. Growth in our power business unit is driven by 2 major technologies. Our 0.18 Micron BCD technology serving power management, audio and display driver products for consumer and enterprise markets as well as our 700-volt technology serving commercial LED lighting and motor driver markets. The market for 0.18 BCD is very large, as the technology can address power management, audio and display driver ICs for nearly all consumer devices for mobile to computing, but is also well suited for more demanding applications in industrial and automotive. Today, much of this market is owned by integrated device makers, but more fab-less and fab-like consumers are engaging with us and finding market success, which, in turn fuels our revenue growth and strong design win pipeline. The 700-volt technology is well suited to the emerging LED market and will grow as this market continues to take hold. The overall market is indeed growing rapidly, as most industrialized countries are phasing out of traditional light bulbs in favor of LED lights. This trend is expected to accelerate in the coming years, presenting strong market opportunity. Today, LED bulbs are largely made up of discrete components, many of which, can be integrated with our 700-volt process and, hence, be provided at a reduced cost. We have production volume in our initial 6-inch based process and are now releasing an 8-inch based process that will enable dye shrink and cost benefit to move more of the discrete component suppliers to our integrated solutions and fuel future growth in this market. As you know, our RF high-precision analog business unit growth is driven primarily by 2 communication markets: the wireless market and the network infrastructure market. The wireless market included wireless connectivity and consumer products, such as mobile phones and tablets as well as wireless connectivity for the Internet of Things, such as the connectivity found in appliances or industrial equipment. Within this market, we produced the front-end RF components that control the transmission and reception of signals. Global Front-End Module revenues are expected to grow by 17% between 2014 and 2015. The combination of strong market growth and our strong market share gains has contributed significantly to our overall growth this year, and this is expected to continue for the next several years, as these trends continue to play out. Smartphone growth in 2014 has been very strong and is expected to continue into '15. We serve this market through components for the RF front-end module, which outpaced the growth of smartphones, since each of these smartphone requires more front-end module units than older phones. They also benefit from significant market share gains, as switches and tuners underwent a technology change from pHEMPT gallium arsenide to silicon-on-insulator platforms, a market in which we have a leadership technological position. We expect this trend to continue in 2015, and longer term, we see the move of the power amplifier from gallium arsenide to silicon germanium technology, as a move that will provide future and further growth opportunities, providing a large serviceable market, that, today is not yet serviced by us. Our technology meets the market needs for lower-loss, high-end linearity switches to support requirements for lower power components from handset manufacturers and higher data rate offerings from network providers. The Internet of Things market is also growing quickly, as everything in our lives becomes increasingly interconnected. We already participate in this market within the RF segment, supporting several communication protocols used in many things already being connected to the Internet. In addition to wireless communication, longer term we expect to participate in the manufacture of many sensors, for example, those built with MEMS or other specialty technologies. The network infrastructure market is characterized typically by lower volumes, but by premium selling prices for high performance and reliability. Within this market, we built high data rate receivers and transmitters for fiber optic links using our high-performance silicon germanium technology. These fiber optic links are used in the network backbone that carries most of the world's voice and data traffic as well as the perimeter of the network, connecting wireless base stations, data centers and even connecting homes directly through fiber-to-home providers. In 2014, this market grew significantly, both from network growth itself and from growth within data centers to support increased data, driven by strong increase in picture, video transmission through the Internet. This trend is well expected to continue through 2015. Longer term, we see growth markets for the same high-speed silicon germanium technology and automotive radar, where we have already some key design wins, which should move to significant revenues in 2016 and beyond. Finally, with regard to the RF high-precision analog business unit, we previously stated that our Migdal Haemek 8-inch factory in Israel was qualified for SOI-based front-end module antenna switch and RF CMOS platforms for power amplifier control, and that qualification was needed to support our overall growing demand from our market leader customers. Eight customers have now designed products in the Migdal Haemek 8-inch factory, with about 40 products presently being manufactured. This has created significant additional revenues for Q4 2014 and will multiply multi-fold over the next year. To summarize, thanks to strong margins from a loyal base of customers within growing markets, we executed the third quarter on multiple fronts. We have fully replaced the $40 million second quarter Nishiwaki revenue contribution with organic growth now manufactured in other cost covered factories. Such activities drove a third quarter three-point margin improvement and has enabled a fourth quarter mid-range guidance of $235 million, a record revenue representing organic growth of 33%. We expect Q4 to realize an additional at least 3 margin points, as the costs in Nishiwaki are brought to a minimum. We entered 2014 in strong position, expecting strong organic customer growth and anticipating the formation of TPSCo. The core growth was realized with a 34% organic year-to-date increase and, hence, replacing the previous revenues from Nishiwaki and replacing them at lower operational costs, and TPSCo has launched as per plan. The first several quarters at TPSCo foundry-type engagements and, especially, the result in projected 2016, 2017 revenues, well surpasses the original business plans presented to and approved by, both the TowerJazz Board of Directors and the Panasonic Board of Directors. As a result of our activities, based upon a strong and loyal customer base, Q4 is guided to be a record-revenue quarter with strong incremental margin growth and will serve as a wonderful springboard into the new year. With that, I'd like to hand the call over to Oren Shirazi, our CFO. Oren, please.

Thank you, Russell. I will now start by reviewing the financial highlights, and later, we'll analyze, in detail, the P&L results and the balance sheet for the third quarter of 2014. In terms of our P&L metrics as compared to the third quarter of 2013, we increased our revenues by 70% from $133 million to $226 million, more than doubled our non-GAAP net profit from $12 million to $31 million, more than doubled our non-GAAP earnings per share from $0.26 per share to $0.58 per share count, improved our EBITDA by 79% from $21 million to $37 million, and recorded net profit under GAAP of $3.6 million for the 9 months ended September 30, 2014. In addition, we strengthened our balance sheet by increasing cash, improving balance sheet ratios and reducing debt maturities, including the $111 million term loan refinancing recently announced. Our increased cash balance went up to $195 million cash on hand, with the generation of $47 million net positive cash flow from operations for the third quarter of 2014, and $97 million positive for the 9 months just ended. Analyzing the P&L in more debt -- depth, starting with revenue, we had a 34% year-over-year revenue increase in organic growth from our top 10 customers. This, together with the Panasonic revenue, caused the significant improvement in our revenue and margins, and more than fully covered the $40 million revenue recorded from Nishiwaki fab in the previous quarter before its cessation. It is important to note that since our revenue from Panasonic are denominated in the Japanese yen currency, and since the exchange rate against the dollar has increased from 100 in the previous quarter to 115 currently, our revenue for Q3 and going forward would have been higher by such a percentage change in the yen currency. However, a similar effect in dollars would have been recorded in our costs in our 3 Japanese fabs, resulting in that the changes in the Japanese yen are naturally hedged by us, given our Panasonic revenue multiyear planned committed contract and the cost in Japan. In other words, the yen currency changes can impact our revenue, but have minimal, if any, effect on our EBITDA growth, operating and net profit. Revenues for the 2014 9-month period were $593 million, up 60% year-over-year, with organic revenue growth of approximately 34% for the company's top 10 customers, excluding Micron and Panasonic, and 27% growth for the top 5 customers. For the 2014 9 months period, gross and operating profit on a non-GAAP basis were $175 million and $98 million, respectively, an increase of 49% and 57% as compared to the same period in 2013. Non-GAAP net profit for the 2014 9-month period was $81 million or $1.61 per share, significantly higher than the $37 million or $1.01 earning per share in 2013. Our non-GAAP gross profit for the quarter was $68 million, reflecting 30% gross margin. This $68 million gross profit reflects an increase of 72% compared to the $39 million reported in the third quarter of 2013, and an increase of 9% as compared to $62 million in the previous quarter, reflecting 27% gross margin. Quarter-over-quarter margin has improved, though revenues have slightly decreased. This is because the reduced Nishiwaki revenue at reduced margins, but they were replaced by 50% incremental margin of our organic growth. Our EBITDA, which is akin to non-GAAP operating profit, was approximately $37 million, an increase of 79% as compared to $21 million in the third quarter, and an increase of 12% as compared to $33 million in the previous quarter. It is important to note that R&D and other operating expenses increased from $18 million in the third quarter of last year to $30 million now, with the growth primarily in R&D expenses associated with the R&D activities in the newly established TPSCo. All this is in order to grab foundry customer engagements and to utilize the available capacity, which TPSCo can manufacture. On a non-GAAP basis, net profit for the quarter was $31 million, which represents 153% growth over the third quarter of 2013. This reflects $0.58 per share, basic earnings per share for the quarter, which is substantially higher than the $0.26 per share reported for the same quarter last year. On a U.S. GAAP basis, the net loss for the quarter was $19 million, representing $0.37 per share, an improvement, as compared to a net loss of $32 million or $0.68 per share in the third quarter of 2013. I note that under IFRS calculation, IFRS are the International Financing Reporting Standards applicable to non-U.S. based companies, our results would have been even better, with net profitability being $8 million better than under U.S. GAAP, due to lower financing expenses under IFRS. In terms of our balance sheet highlights, as of the end of the third quarter, we continued to improve our balance sheet ratio's debt-to-EBITDA ratio and current ratio. Cash and short-term deposits as of September 30, 2014, were $195 million, an increase of $72 million when compared to $123 million as of the start of the year. The increase in cash during the first 9 months of 2014 was attributed mainly to $118 million of cash generated from operating activities, excluding interest payments of $21 million. In addition, during the 9 months period, we invested $73 million in fixed assets net. We repaid $35 million of debt principal. And we received $14 million proceeds from exercise of options and bonds issuance. In addition, we included in the cash balances $58 million in cash resulting from the TPSCo transaction and received a loan of $86 million from JA Mitsui and Bank of Tokyo, which has been used solely to repay a bridge loan granted from Panasonic earlier in connection with the establishment of TPSCo. In addition, funds received from Nishiwaki asset sale, net of Japanese employee retirement payment, amounted to $12 million. In October, we signed an agreement to refinance $111 million of our existing bank debt during the next 2 years, with the loan maturing in 2018, thereby, we reduced 2015 and 2016 maturities from $101 million to only $24 million. Finally, we're happy to report that following the increase in our market cap and our average trading volumes, MSCI announced that it intends to help Tower to its index -- in its next index review planned at the end of this month. We look forward to the additional recognition that this will bring to us amongst international institutional analysts, investors, EPS and other investment funds. This ends my financial review. And I will be happy to answer questions at the end of the conference call. And now I wish to turn the call to Limor. Limor?

Thank you, Oren. Before we'll open up the call to the Q&A session, I would like now to add the general and legal statements to our results in regards to statements made and to be made during this call. Please note that the third quarter of 2014 financial results have been prepared in accordance with the U.S. GAAP and the financial tables in today's earnings release includes financial information that may be considered non-GAAP financial measures under Regulation G, and related reporting requirement as established by the Securities and Exchange Commission as they apply to our company. Mainly, this release also presents financial data, which is reconciled as indicated by the footnotes below the table on a non-GAAP basis after deducting depreciation and amortization, compensation expenses in respect to options grants, and finance expenses net, other than interest accrued, such that non-GAAP financial expenses net, include only interest accrued during the reported period. Non-GAAP financial measures should be evaluated in conjunction with, and are not a substitute for GAAP financial measures. The tables also contained the comparable GAAP financial measures to the non-GAAP financial measures, as well as the reconciliation between the non-GAAP financial measures and the most comparable GAAP financial measures. EBITDA as presented is defined in our quarterly financial release. EBITDA is not a required GAAP financial measure and may not be comparable to a similarly titled measure employed by other companies. EBITDA and the non-GAAP financial information presented herein should not be considered in isolation or as a substitute for operating income, net income or loss, cash flow provided by operating, investing and financing activities, sales share data or other income of cash flow statement data prepared in accordance with GAAP and is not necessarily consistent with the non-GAAP data presented in the previous filings. I would like now to turn the call over to the operator. Operator?

[Operator Instructions] The first question is from Cody Acree of Ascendiant Capital.

Russell, maybe, if we could start with your RF business just as strong as that has been and the numbers we've been hearing from all of your customers. Some have even pointed to some SOI constraints. Just maybe, if you could talk a little bit about the strength of that market, and your capacity additions. And are you seeing any constraints right now to supply?

So as far as your -- part of your question on the strength of the market itself. We see, really, very, very big strength in the predominant reason of the cross qualification of the Migdal Haemek fab 2 was for the fact of needing additional capacity beyond what we could be supplying in Newport Beach, and that's incremental to having added capacity in Newport Beach so that the market strong. I had stated in the call that the revenues themselves for this -- these technologies in Fab 2, we expect a multi-fold increase of that from Q4, which is already substantial. So it's growing very, very strong. As far as supply constraints, there was, indeed, toward the end of Q3 a constraint on SOI substrates themselves. That has since been alleviated by at least our predominant supplier and additional suppliers have commenced on the market as well. But I don't think that the substrate itself should be a problem going into 2015 and throughout 2015. But certainly, there was a period of time in Q3 that the supply was strongly constrained.

And Russell, as far as your market share versus IBM, you've been taking some significant share. How much of that, I guess, is factoring in? And I guess, do you have visibility as to what is happening with underlying demand versus your driver of just share gains?

I'm not sure I fully understood the question, Cody. Please, one more time.

I guess, what I'm trying to get at is, the demand growth that you're seeing, can you parse that between what you're seeing as the underlying health of demand versus your market share gains against IBM?

I think, certainly, that the demand is increasing. There's no question there. But if you look at 2013, we had low single-digit revenue from SOI. And this year, it'll be multiple tens of millions. So intrinsically, as IBM was the predominant competition, we've gained market share against IBM. How to quantify market share gains versus an increase in the overall SOI demand, that's a little bit hard to do. But if you come from nothing to enough many tens of millions of dollars of revenue, certainly, the market share has grown. And the demand is higher going into 2015 than we see in '14. I believe that for many of our customers, all new platforms are being designed to us and taped out to us. So I would expect that our market share will continue to grow. But I really -- I don't know how exactly to quantify the question or the answer, I mean. Sorry.

And Russell, your customers have been putting out -- your RF customers have been putting out some extremely strong quarters, and there have been some concerns of, maybe, overheating in that space. I guess, what's your visibility, knowing you're a couple, 3 steps removed from the enhanced set market? But -- just your comfort in the stability of that market going forward.

I think it's very strong. I think that the amount of application that's going on smartphones. The usage of smartphones, the usage of tablets, certainly, has cannibalized a big portion of the PC market. And I don't see that declining the functionality of smartphones, the necessity of more front-end modules is not declining. So I think it's a strong market. Will there be 1 quarter, at some point, of a pullback in the overall market? It's maybe possible. But the market itself is growing and will continue to grow. The question is to be diversified enough in our customer base and for our major customers to be diversified in their customer base. I mean, it's -- from the end, cell phone makers, there's a handful of the largest and 2 handfuls that make up 85%, 90% of it. If you have a customer that has bolstered its market based on 1 provider and that provider loses share to someone else then, there's a risk there. But I think for the market, there's not a risk. And overall, company S, lose share to company L, who might lose share to company A, maybe. But as long as you're diversified enough and your customers are diversified enough, I don't think that it's a problem for us.

Perfect. And then, Oren, just a couple of things for you, on the gross margin side. Can you maybe, give us any color on how much of the margin lift was from Nishiwaki's closure versus just your organic better, richer product mix? And then, how that plays out into the December quarter of margin improvement as well.

Yes, sure. So you'll see here that the gross margin went up from 27% to 30%, which, this quarter Q3 includes about $6 million, our last type of expenses in Nishiwaki. We still have, like I mentioned in the script, we are in the process of selling some assets, which are funding in full, plus some excess cash should funding full, plus some excess cash for the employee pension liabilities. In order to still operate the facility and still have there -- I mean, it's not manufacturing, but still the machines are there, most of them. So we do include some electricity costs, some payroll for a few employees that are there in order to deal with the process of de-installation of the tools. So total in Q3, we include such expenses of $6 million. Of course, it's good, because it's compared to $30 million fixed cost that we had quarter before. And going forward, $6 million, I think, you can assume that for Q4, it will be maybe, $2 million, and from Q1 '15, it will be nothing. So really will support the continued improvement of the margin.

And when you look at the ForEx change, you gave us some color on the absolute from 1 to 115. How much of that impacted Q3? If we just look at it apples-to-apples revenue, what was the impact in Q3, and what's the impact in Q4?

So it goes by the average, pretty much by the average, because it goes by each shipment of wafers. So Q3 average, from our point of view, was 100 -- 108 or 1 0 8, so 8%. And therefore, Q4, so far, now they ended at 115. In October, it was more like 110. So expected to be, if the yen rate will stay like that 115, expected to be at about 113.

And then lastly, just a split of expenses as we pull, not only the revenue out, but expenses out for the Forex impact. How does that split between cost of goods and your operating expenses?

What do you mean, ForEx?

Well, when you're talking about this currency translation, we're backing out revenue, but we're also backing out some expenses. So how do we -- going to look at modeling, how do we look at the breakdown of those expenses between COGS and operating expenses?

So about $10 million is in the OpEx from the entire expenses associated with the TPSCo, and all the rest is in the COGS.

The next question is from Jay Srivatsa of Chardan Capital Markets.

Russell, you mentioned a high-level trend in your transcript about how IDMs are looking to go fab-light, or potentially, fab-less. Could you expand on that? And help us understand what is precipitating such a move, and how quickly does that play out in terms of Tower benefiting from further demand shift.

Certainly. So there was a press release, what, 6 weeks ago, 7 weeks ago from Fairchild, about Fairchild's ceasing operations of both packaging and manufacturing facilities. That's one example of integrated device makers, who, possibly, have factories that are not being optimally utilized and would be then consolidating to move certain flows into a foundry. And hence, the foundry can benefit from incremental utilization. And the IDM would benefit from not covering unused fixed cost. And that's, again, if that was Fairchild's reason or not, that, I couldn't say, but it's one such recent example of a very well-known IDM, that is ceasing operations. You have older examples happening fairly continuously. The fab light model has happened for a long time and it continues very much so. If you look at power management, at this point, the core of all power management is being produced at 0.35 micron. As you have power management, development and activities at 0.18, 0.13 micron, would one of these IDMs that has a very strong market position, but factories that are no longer state-of-the-art, would they invest for increased capability, whereas, they know that the first years of that capability will not be highly utilized? Or would they invest with the foundry to develop their flows at the foundry, and, hence, not have an underutilized factory, as the flow themselves was gaining market and market share? So that's the fab light drive, and the fab light has been -- I mean, that's the whole reason, that foundries themselves have grown. But where you have IDMs that themselves own very good intellectual property in the flow, that's where they would transfer the flow into our TOPS model that we have, TOPS means transfer optimization process services. And we have a specific team that works with the integrated device maker to take their flows and bring them into our factories. In other cases, you have integrated device makers that design to the flow of the factory itself or the flow of the foundry itself. So you have 2 different models, but 1 model is, very specifically, someone who is very strong in their own flow and wants to do a development or needs additional capacity for an existing technology note, instead of investing for that capacity themselves, thinks it's a win-win. The foundry, I'm going to give them incremental wafers that enable them to have a steady-state higher utilization. And I don't need to invest CapEx or infrastructure that I might not be utilizing especially immediately. Did that answer your question, Jay?

Yes, it did. Thanks for the detailed explanation. Shifting to the JV, with that Panasonic JV in place, are you now better engaged with Japanese customers? And when do you expect to be able to add to your customer base from the local customers in Japan?

I think the answer's a very strong and resounding yes. A name card with Panasonic on it certainly helps a salesperson to have credibility in Japan. And the initial answer, I mean, we have several Japanese customers that have come to TPSCo for a pure foundry service. Also, I had mentioned activities within our TOPS business group, we have just more or less finalized an agreement with a Japanese integrated device maker for transferring their flow into TPSCo itself. I think that this specific win that we just had would have been very difficult to have had without the Panasonic name in Japan.

All right. And then, the last question on the India project. Where are things now? And when do you expect to be able to report further progress there?

Things right now are pretty much unchanged. We submitted the -- what was it, a year ago, plus, minus. India government came back and said that they were going to make it into a 3-tier approval and each one having its own name. We received the first tier approval, to which, we had to respond, which we did. We received the second tier approval, to which, we had to respond, and we did. The third approval is what we're waiting for. There's no reason to think that it won't be granted. When are we expecting it? I believe that the government itself had said that they'll complete the final approval process before the end of the year. That's what the government said. The most important thing about the consortium that we have with Jaypee and IBM is that this is not an intrinsic part of our business plans or business model or a multiyear plan. Basically, if it happens, it's a very, very nice incremental piece of business for us. It's not, per se, used revenue, but it's very, very high margin revenue. And if you just look at the fact that the activity that we'll be doing is very close to 100% margin. And if you're looking at any yearly tens of millions of revenue that's at 100% margin, it's very accretive. But it's not part of our multiyear plan. It certainly wasn't part of our annual operating plan. And depending on what we find out before the end of 2014, it most likely, unless it's approved, we would not include it into our 2015 annual operating plan. But if it does happen and when it happens, it will be a very nice accretive point of margin dollars.

Next question is from Lisa Thompson of Zacks Investment Research.

I just have a couple of quick questions. One is kind of philosophical and the other is just a housecleaning. First off, on the -- when you go in on the joint venture with Panasonic, do you find that your customers are hesitant to go to that business, because they compete with Panasonic? Or are you finding that more of the competitors are more amenable because you're in the mix? And have you seen any people that compete with Panasonic coming to you?

It's a very interesting question, and a good question. We have not, to date, had any customer be fearful of doing something in TPSCo, because of stating that Panasonic competes with them. For the most part, the Panasonic demand within the TPSCo factories, presently and in the past, had not been to put semiconductors on the market, but integrate them into systems. Most of our customers are not system makers. In fact, very few of our customers are system makers. So our customers having a fear of Panasonic competing with them is not -- it's not an apple-to-apple comparison in most cases. Panasonic is 1 level up as really a system provider. Most of our customers are chip providers. So that's, maybe, the philosophical answer. The real tactical answer is, no, we've not seen anyone that was fearful. And maybe, to the opposite, when you deal with, in particular, Korean customers, they're very, very thrilled about working in Japan, because of the brand of Japanese quality and, in specific, Panasonic has an amazing reputation within its facilities. Does that answer your question?

Yes, that's great. I was curious about that. And the other -- my other question is just talk about the interest expense. I know you've had a lot of moving parts on loans and debts. And I thought it would be down a little bit more than it is this quarter. What is interest expense going to look like next quarter and then next year? Has anything changed to change the numbers?

Yes, so we have, actually, have 2 lines for the interest expense. The first line is real interest expense, meaning interest which is payable. And that line is pretty stable, it goes down a little bit. Currently, it's $8 million a quarter. And all the recent announcement has not too much effect on it, a slight reduction, but not too much effect, because we had those Jazz bonds extension, and we had that refinancing with term loan, but we kept the same amount of interest. So it should be -- continue to be flat like that, $8 million in the coming quarter, and also, the same assumption is relevant for 2015. From 2016 and below and above -- from 2016 and after that, it will be lower, because some debt vehicles are redeemed during the next year. So from 2016, it will go down to about $6 million, and later, $5 million a quarter.

Okay. So for next year, taking $8 million times 4, or it sounds about right.

Yes. Yes.

Our next question is Eric Hall [ph] of Cole Capital.

Two questions. Regarding the Panasonic joint venture in Japan. Can you kind of clarify how much non-Panasonic revenue you're going to be able to add in 2015? I know you talked about later years, but do you see any revenue for -- that you signed up for -- that will be actually recorded in 2015 that's not Panasonic related?

Yes. I believe, we'll start seeing some revenue that's non-Panasonic related at the end of Q1 or Q2, and that will continue to grow throughout '15. But the major portions of the revenue increases will come in, in the very end of '15, and then, ramp in '16.

Okay. So I mean, you don't have to comment if you're not comfortable, but would $25 million of revenue incremental to the Panasonic guarantee in 2015 be reasonable?

I would think so.

Okay. And then, a second question, kind of an important one, on free cash flow, since it's going to be looking so good going forward. Could you please clarify what -- going forward next -- for 2015 or the next 12 months, what is depreciation and amortization going to be? What are your capital expenditure plans? And do you see your working capital being a source of cash or an area we are investing?

Okay. So in regards to working capital in our industry, in our type of foundry business, usually, typically, the customer payment terms is about 30 to 45 days and the vendors is about 90 to 120 days. So even in a period that we are growing, and now, it's such a period, as we are growing, significantly, you'll see that there is no negative effect to the increased working capital on the cash flows. You'll see, for example, that Q3, which was the best in terms of the most ramped one, the most growth one, still, we have a positive cash flow amortizations[ph] , about $50 million, which is the highest quarter of the year, and also, this year is very good. So not negatively affected from that at all, because of what I mentioned that, actually, customer payment terms are better than -- or shorter than vendor payment terms. In terms of depreciation and amortization and CapEx, we don't give the exact focus for that amount. But you can see Q3 figures, and pretty much, we don't expect a material difference, as I can relate to it in specifics, depreciation in the third quarter. As you can see in the non-GAAP report in the adjustments call, it's about $53 million. So this should go gradually down, it will be only lower, it will not be higher. Most of that is historical result of the expensive cost of building, establishing fab 2 and ramping it up. So it's going down, because the depreciation schedule is at its end period. So this should go down to the $40 million quarter level along the second half of 2015 and then in '16, even lower. And the CapEx that we reported on that, that the CapEx were $73 million in the year-to-date so far. So about $25 million per quarter, this should be the sustainable run rate of the CapEx payment.

The next question is from David Duley of Steelhead.

A question on the revenue that you can bring into the new Panasonic fabs that you've -- in joint venture. I guess, it's about $350 million to $400 million of total capacity. How long would you think it would take you, under ideal circumstances, to fill the factory?

I think, we're in pretty ideal circumstances right now. And I believe that we'll, probably in the 2017 timeframe, pretty much use all of the open capacity there, at least by run rate in the second half of 2017.

Okay. And when you look at your core business and the segments that you talked about, which area do you think it's going to sop up the most incremental capacity? Do you think it's the RF power amplifier and funding guys or analog power management? Just give us an idea what you think is going to absorb the most capacity.

Capacity of TPSCo or just capacity of the company?

Which one of your segments of your current customer base is going to use your incremental capacity that you have in our joint venture? Is it going to be the RF guys, power management? Which area is going to use most of the current capacity that you have to offer them?

I believe that will be in the high-end side. The image sensor, the cameras, that will be at the 65 nanometer. It will be in the IDM transfer, the TOPS business and in power management. So those are the units that we'll be using for the TPSCo. As far as the front-end module work that we're doing, we are not transferring that to TPSCo. That's where we had qualified the Migdal Haemek fab 2, and we'll continue to do that both in Newport Beach and in fab 2. The high-end optical transceivers, the silicon germanium platforms are more nigh near impossible to transfer the begin with. It's very, very hard to match the output parameters, and the major drive that we have is to allow everything, other than the high-end silicon germanium, to be able to be sourced in Migdal Haemek, and hence, to continue to grow the high-end silicon germanium business and meet customer needs in Newport Beach. There is one additional that I should say to that. We are qualifying a 300-millimeter SOI platform in the Uozu factory at TPSCo. But I do not believe that, that will ramp into any high volumes within the next couple of years. But we want to have a 300-millimeter platform available for customers as a roadmap for the customers on longer term. But the Front-End Module work that we're doing is not targeted to be using capacity at TPSCo.

Okay. And help me understand with the gross margin profile of the incremental business that you bring into joint venture. What would you expect that to be? And is it higher than your core gross margins?

Okay. So currently, we have -- and I'll also, maybe, relate to what you said in the beginning, you mentioned $350 million number. So currently, what we announced in past is that the Panasonic contract is between $90 million to $105 million revenue a quarter, and here I put a footnote in regards to what I said in the beginning, all depends on the yen. So if the yen goes up 10%, so it's between $80 million to $95 million, et cetera. But let's assume the $90 million to $105 million a quarter, which is $360 million to $420 million a year, that's from the Panasonic committed contract. And of course, since it is a committed contract, the gross margins are not the best gross margins that we have in the company. However, any incremental revenue business that we bring into the factory is from third-party business, from all these power RF and all that segments that we bring, SIM, CMOS image sensor is, of course, it's very high incremental margin, because all the fixed costs are actually already covered by this business model with Panasonic. So basically, upon full utilization, you should expect that the margins, the gross margins in the TPSCo fab will be, for sure not lower, but possibly, very much higher than -- pretty much higher than the standard gross margin you'll see in our business, because of the efficiency of those fabs, because of the fact that we have one of the most advanced fab. The most advanced fab that we have, we have in the TPSCo, which is the Uozu fab at the 65 and 45 nanometer.

And the incremental margins we should expect from that business, is that a 60% or 65% kind of number? And the reason it's so high versus core gross margins is because you got the Panasonic fixed [ph] cost coverage. Is that correct?

Exactly. Exactly.

Okay. And just a couple of clarifications from me. I'm somewhat new to the story, but it struck me as very positive that in this joint venture, you have a 300-millimeter, 65-nanometer advanced analog factory. Could you talk about, if you were to greenfield that fab, what possibly would it cost you? And what are you going -- that seems like a really very productive and kind of a gem of an asset. Could you just talk a little bit more about why that's a strategic asset?

Yes, on the first part of the question, I think it would be in the neighborhood of $2 billion, if we were to greenfield the facility, and that's minimum. It could be substantially higher than that depending on the capacity we would be building in it, but it would be a minimum of $2 billion. The strategic part of it is twofold. It begins with the fact that Panasonic, within their core capabilities, had an extremely effective quantum efficiency, bright pixel, dark current, CMOS image sensor flow at 65 nanometers. As far as the figures of merit, it's probably the only flow in the world that honestly competes head-to-head with Sony. So strategically, to have now capability of doing small pixels in a 300-millimeter wafer at 65-nanometer, with the core competence and capability that was built into that factory for image sensor, that's why we were able to get the press release, the activity with Himax. And we are really in negotiations there with very, very big people at this point for further capacity usage in the 300-millimeter facility for image sensors. So that the first and foremost strategic point was having an avenue for image sensors, that we did not serve ourselves previously. We are very strong within gesture control, very strong in industrial, very strong with studio cameras, but not in very small pixel, and unbelievably, strong quantum efficiency capabilities. Secondly, is the fact of having -- I had mentioned it before, but a roadmap for customers that, for whatever reasons, might, at some point, want to use 300-millimeter wafer size. So to be able to develop that now, I mean, we have had opportunity in the past to acquire a 300-millimeter manufacturing facility at a relatively low cost. The difficulty was that we didn't have the backup business to fill the factory at such a point when a loading agreement would run out. And the running cost of a 300-millimeter factory is quite high. Any factory, it's high, but 300-millimeter at 65-nanometer, 45-nanometer technologies, it wasn't within our existing customer base. And hence, to buy a factory, however, low the cost, have a loading agreement that would maybe start decreasing after 4 quarters, 6 quarters, but know that to build up a new customer base in those technologies would take us 3 years, would be a very big hole in the finances of the company. So strategically, to have a 300-millimeter factory of reasonable capacity that we can vet and build a customer base within, when the next opportunity would come about, if needed, in 3 years, 4 years, 2 years, whatever it might be, to grow in 300-millimeter capability, we already have the customer base in the flows qualified. So at that time, if an opportunity is there, you can take the opportunity with a strong belief that you won't go for 8 to 12 quarters at a strong quarterly cash loss through the operations of that factory. And that was, maybe, too long of an answer, but did it make sense? I mean...

Absolutely. I guess, in summary, the backstop of the fixed cost coverage from Panasonic is going to allow you to offer your customers a 300-millimeter process, which has got to be a huge cost advantage for some of your key customers.

Correct.

There are no further questions at this time. Mr. Ellwanger, would you like to make a concluding statement?

Certainly. As always, we really do thank our customers for their trust in us as a long-term partner. At the organic growth that we're having, that we've demonstrated Q1 through Q3 of 34%, I think that, that is demonstrative of having the right customers and doing the right things for the right customers. And we look forward to seeing 2015 and growing that relationship and growing out capacity and utilization within that relationship. We thank, very much, our investors for their belief in our management and in our business model, short term, midterm, long term. Certainly, we thank our employees for their capabilities, dedication and passion. And that, really, is what has driven us to be the #1 specialty foundry in the world. We really are at a very, very exciting point in the company's life, a point of big organic growth, big opportunities. And now it's really in the hands of operational execution and continuing to drive roadmaps, that we're aligned with key customers on their next-generation and 2-generation out needs. So thank you for your interest. And just a last statement, on December 2, we'll be presenting at the LD Micro Conference in Los Angeles. In mid-January, we'll be presenting at the Needham Conference in New York. I'll be pleased to meet any of you at either of those conferences. And anyone who would like to have a follow-up call with either Oren or myself, we'd be very amenable to that as well. So thank you for interest. It was really a good quarter, and I think Q4 promises to be better. Thank you.